Data analysis systems typically provide various types of graphical interfaces and visualizations used to display data identified based on execution of search queries. Using these interfaces, for example, a user typically can create search queries including various functions, arguments, and clauses to identify data of interest to the user. As one example, a sales manager might create and execute a query to search a sales information database for data corresponding to product sales within the past year. Based on execution of the query, a data analysis system might generate a graphical interface visualizing the results data using a bar chart, pie chart, or other type of visualization.
Some data analysis systems further enable users to “split” search results data by specifying in a query one or more fields upon which to split the data. For example, referring again to the example of a query searching a sales information database, a user might create a query to search for data relating to product sales in the past year. Each entry in the database might contain various fields including a type of product sold, a region where the corresponding sale took place, and a dollar amount associated with the sale. In this example, a user might desire to split the sales data by region so that the user can compare a total sales amount across various regions. A resulting visualization may display a bar chart, for example, with one axis of the chart corresponding to the various regions, and a second axis of the chart corresponding to a range of total sales amounts. In this manner, the user can compare a relative height of each bar in the chart to compare an amount of total sales occurring in each region.